RE: altglyph

Paul,

At 11:42 AM 9/23/99 -0700, Paul Topping wrote:
>responses below.
>
>> -----Original Message-----
>> From: Jon Ferraiolo [mailto:jferraio@Adobe.COM]
>> Sent: Wednesday, September 22, 1999 1:51 PM
>> To: Paul Topping
>> Cc: 'www-svg@w3.org'
>> Subject: Re: altglyph
>> 
>> 
>> Hi Paul,
>> 
>> At 09:08 AM 9/22/99 -0700, Paul Topping wrote:
>> >Hello,
>> >
>> >I've been reading Section 12.6, "Ligatures and alternate 
>> glyphs", of the SVG
>> >spec where the "altglyph" attribute is described. I'm a 
>> little confused as
>> >to the details of altglyph.
>> 
>> In the August 12 spec, 'altglyph' is a CSS property, not an attribute.
>> Since it is a property, it can be specified as part of an 
>> external style
>> sheet, as part of an internal style sheet via a <style> 
>> element, or within
>> an inline style via the 'style=' attribute, which is available on most
>> elements in SVG.
>
>I was unable to find any mention of "altglyph" in any of the CSS docs. Could
>you point me at some documentation on it?

"altglyph" is a new thing which will be part of other web standards down
the road (or so it appears, given the involvement of the I18N folks), but
for now SVG is taking leadership.

I believe the CSS working group has had discussions about this in
conjunction with CSS3 and that they are waiting to see how it shakes out
with SVG.

>
>> The international working group has a strong interest in 
>> 'altglyph' as it
>> has considerable utility in a variety of scenarios, including 
>> Japanese.
>> Also, 'altglyph' has applicability to other XML grammars such 
>> as XHTML.
>> They have given feedback to the SVG working group that 
>> questions whether
>> CSS properties are the best approach for 'altglyph'. There is 
>> a chance that
>> 'altglyph' could change from a property to an element or an attribute.
>
>I think "altglyph" has applicability to math rendering as well. If we render
>MathML by converting it into CSS (possibly with new math-related CSS
>elements) and HTML, we will rely on plain text for rendering individual math
>characters. Regardless of how many math characters are in Unicode (now or in
>the future), we will be dealing with a lack of font and character knowledge
>in browsers for some time to come. It seems like "altglyph" might be needed,
>at least in the interim, to get the desired character in the desired math
>font to appear on the page.

Yes, I think you are correct that "altglyph" might have utility with MathML.

Another SVG feature which you should look at is SVG fonts (chapter 13). I
would expect that CSS will add SVG fonts to their list of allowable web
fonts. Also, because SVG fonts are XML, they can be added inline within
another grammar using XML namespaces.

>
>> >- It is not clear to me what element this is an attribute 
>> of. Is it the
>> >"text" element? I couldn't find altglyph mentioned in the DTD.
>> 
>> With the August 12 spec, because 'altglyph' is a property, it can be
>> specified in all of the ways I have described earlier. 
>> However, it only
>> makes sense to specify it on a "text" element.
>
>I know this isn't your responsibility, but do you know if any of the SVG
>prototype renderers deal with "altglyph" right now?

I doubt if any SVG renderers have implemented "altglyph" yet. Adobe's
implementation hasn't gotten to "altglyph" at this time but we will support
it by the time we make our final versions of SVG browser plugins available
to the world at large.

Jon Ferraiolo
SVG editor
Adobe Systems Incorporated

Received on Thursday, 23 September 1999 15:50:44 UTC